We contribute Policy Reuse as a technique to improve a reinforcement learning agent with guidance from past learned similar policies. Our method relies on using the past policies ...
This paper presents a Path Swapping (PS) method which enables to enhance the security of Quasi Delay Insensitive Asynchronous Circuits against Power Analysis (PA) attack. This appr...
We propose a novel Bayesian learning framework of hierarchical mixture model by incorporating prior hierarchical knowledge into concept representations of multi-level concept struc...
For years, the cryptographic community has searched for good nonlinear functions. Bent functions, almost perfect nonlinear functions, and similar constructions have been suggested ...
Estimation of Distribution Algorithms (EDAs) are a class of evolutionary algorithms that use machine learning techniques to solve optimization problems. Machine learning is used t...